Statistical Mathematics

Statistical and mathematical advice and analysis is vailable to industrial, commercial, academic and institutional clients. It also provides short courses in statistical thinking and practice, tailored according to customer requirements.

To support automotive design, research, development and production, we provide expertise in all kinds of statistical and mathematical modelling and analysis. These include, for example, computer experiments, data analysis, decision analysis, experimental design, fluid-flow problems, forecasting and dynamic models, graphical models, multivariate analysis, quality improvement, reliability analysis, maintenance scheduling, software for numerical computation, software testing, and spatial statistics.

Of particular relevance to the automotive industry are experimental design techniques (including Taguchi methods) and Bayesian statistics. Experimental design is a statistical technique which allows the very efficient investigation and prediction of processes which are influenced by large numbers of possibly interacting factors, such as engine and aerodynamic performance. Bayesian statistics is concerned with the combination of expert knowledge and statistical information to analyse the uncertainties in large and complex problems, for purposes such as decision making and forecasting, and is especially appropriate whenever experimental data is expensive or difficult to obtain.